Catena 65 (2006) 80 86 www.elsevier.

com/locate/catena

Phosphorus content in five representative landscape units of the Lomas de Arequipa (Atacama Desert-Peru)
Andre Fabre a,*, Thierry Gauquelin a, Francisco Vilasante b, Aldo Ortega b, Henri Puig a
b a Laboratoire Dynamique de la Biodiversite, 29 Rue Jeanne Marvig, 31055 Toulouse Cedex, France Universidad Nacional San Augustin, Instituto Regional de Ciencias Ambientales, Casilla 985, Arequipa, Peru

Received 27 September 2004; received in revised form 21 September 2005; accepted 12 October 2005

Abstract Phosphorus forms and content were studied in soils of the Lomas de Arequipa (Atacama desert, Peru) using a fractionation method. These Lomas are small hills periodically submitted to the El Nino-Southern Oscillation (ENSO) which causes heavy rainfall. Sample soils were randomly selected in five landscape types characterized by vegetation: cactaceae (Cac), cactaceae and herbaceous (CacHerb), shrubs (Shr), trees with cover <60% (Tree) and shrubs or trees with cover > 60%) (ShrTree). All the soils were strongly acidic and classified as loamy sand, sandy loam or silt loam. Organic carbon content was under 1% in Cac or CacHerb, then increased strongly in ShrTree (6.50%). Considering phosphorus, all the forms (labile as well resistant forms) increased markedly from Cac soils to ShrTree soils. In all the soils, the labile forms (Resin-P: range 45 105 Ag g 1; NaHCO3-Pi: 23 123 Ag g 1; or NaHCO3-Po: 10 122 Ag g 1) were very high. These high phosphorus contents were attributed to the specific climatic conditions of the Lomas that feature a long period of vegetation dormancy (very dry period) and a short period of growth, following ENSO-associated precipitation. We suggested that during the dry period, plant decay and microbial cells death lead to release and accumulation of labile P in the soil, the rainfall wetting the soil, permitting vegetation growth. In this respect, the Lomas climatic conditions contribute to soil fertility, especially as labile forms of phosphorus are chiefly concerned. D 2005 Elsevier B.V. All rights reserved.
Keywords: Soils; Phosphorus fractionation; Lomas; Peru; ENSO; Atacama desert

1. Introduction Coastal deserts such as Atacama (Coastal Peruvian desert continued by the Northern Chilean desert) present specific characteristics: a) they are the driest among all deserts; b) the general climate is mild and uniform; c) the temperature is fairly evenly distributed throughout the year; d) they are subject to winter fogs. These climatic conditions impart to coastal arid regions unique characteristics compared to arid regions characterised by high mean and large amplitude temperature. The aridity results from several combined factors, especially the permanent high pressure area over the Pacific Ocean and atmospheric stability induced by the cold northward flowing Humboldt Current. This cold current makes the air become cool or cold but dry and very stable
* Corresponding author. E-mail address: acfabre@aol.com (A. Fabre). 0341-8162/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.catena.2005.10.004

overall, unable to produce precipitation. At the same time, there is very little evaporation and humidity is confined to a low level, giving persistent haze. Whereas mist may occur any time throughout the year, there are some particularly foggy periods, generally at the end of the austral winter and in early spring (Zavala Yupanqui, 1993). Along the Chilean and Peruvian coasts, elevations between 600 and 1000 m are the most favourable for fog formation (Osses McIntyre, 1996). The Atacama desert is strongly affected by El Nino (disruption of the ocean atmosphere system in the Tropical Pacific with consequences for weather around the globe) which generates abundant rainfall. El Nino-Southern Oscil lation (ENSO) is a coupled ocean-atmosphere phenomena that has a worldwide impact on climate. ENSO, which seems to occur with a cyclic rhythm in coastal Peru (every 10 years on average) induces exceptional rainfall in these regions. However, since the nineties, ENSO has occurred every 2 to 7 years. The last very rainy

A. Fabre et al. / Catena 65 (2006) 80 86

81

Fig. 1. Study site location.

periods occurred in 1982, 1992 and 1997 1998. In several parts of the Atacama desert as in the Arequipa region (South Peru), the coast is dominated by low hills (elevation varying from some hundred to about 1200 m) termed Lomas in Spanish (geomorphological sense). The same term refers to the fog caught on these hills (climatic sense) and to the vegetation arising during the foggy season (phytological sense). In the following text, the term Lomas is used in the global sense, comprising all three of these notions. The vegetation is composed of numerous ephemeral but also of perennial species, ligneous plants and cactaceae. Some studies have been published on the Peruvian Lomas (Pefaur, 1982; Ferreyra, 1993). The Lomas are utilized for forage and to gather woody species for fuelwood (Ferreyra, 1977). They are periodically used for grazing livestock (cattle, sheep and goats), especially during ENSO events, and possibly as grazing land during seasonal livestock migration during the Spanish period. Considering the soils of deserts, studies are scarce and mainly concern hot deserts or arid ecosystems (Lajtha, 1988; Lajtha and Schlesinger, 1988; Cross and Schlesinger, 2001). At the moment, no information exists on the soil characteristics of the Atacama desert or of the Lomas. In this paper we consider some general soil characteristics and we emphasize the different forms of phosphorus in soils of five representative vegetation types (Lomas types) of Lomas de Arequipa (South Peru, Fig. 1). Hypothesis of a close relationship between labile phosphorus content in the soils and ENSO events inducing exceptional rainfall is discussed.

71-40V W; 16-90V 17-40 S). In this region, average annual precipitation is only < 50 mm below 500 m alt. and several years may pass without rainfall. The driest period occurs from January February to April. From May to October, heavy fog (relative air humidity near 75%) permits vegetation growth. The average annual temperature is around 18 -C and the annual variation in temperature is small with a minimum of 9 12 -C in July and a maximum of 25 -C in January February (Zavala Yupanqui, 1993). When the coastal topography is flat, the seasonal fog dissipates inland but where isolated hills (150 to 1000 m) intercept the fog, a fog zone appears allowing the development of rich vegetation termed Lomas formations separated by areas without vegetation. In Peru, around 40 Lomas formations exist, among them the Lomas de Mollendo. The bedrock is acid igneous (granodiorite) with local clastic sediments (sand, clay, sandstone or conglomerates). The non-consolidated parent material (particles <2 mm) pertains to the loamy or sandy texture groups. The soils are in the Aridisols class characterized by low organic carbon. 2.2. Soil sampling Using SPOT images (SPOT 661 384; July 1995) and aerial photography, 8 different landscape types were identified in the Arequipa region. Five of them were retained in this study: cover dominated by cacti (Cac), by cacti and herbaceous (CacHerb), by shrubs (Shr), by trees with percentage cover < 60% (Tree), and by shrubs and trees with percentage cover > 60% (ShrTree). The distinction between these landscape types resulted from a site vegetation study (120 sample plots of vegetation statistically analysed using correspondence analysis). Some dominant species are listed in Table 1. In each type, 4 sampling areas were randomly selected using a grid. However, as in desert landscapes, the vegetation strongly influences soil nutrient content, soil samples were randomly selected, after eliminating nearness of vegetation patches. Soil samples were taken in the first 5 cm after discarding the litter when necessary. They were stoney, particularly in Cac and CacHerb. The soil samples were stored for grain size and chemical analysis. Sampling and vegetation studies were performed during September 1997, before a rainy period. 2.3. Chemical analyses Soil samples were analysed for grain size, pH in water, organic carbon using CHN auto analyser, and different forms of phosphorus. Sand, silt and clay percentage were estimated using the pipette method and the soil texture classes were determined using the Soil Science Society of America chart. Total phosphorus was fractionated using a sequential extraction method (Hedley et al., 1982). The sequential extraction removed inorganic P (Pi) and organic P (Po) of increasing chemical stability with different geochemical or ecological significance. First, the most labile inorganic

2. Materials and methods 2.1. Study site The study site was situated near the town of Mollendo, in the Arequipa region, on the south Peruvian coast (72-10

82 Table 1 Floristic composition of each Lomas type Lomas type Cacti Cacti and herbaceous Altitude (range) 160 680 620 790 % Plant cover 13 2 10

A. Fabre et al. / Catena 65 (2006) 80 86

Some dominant species Neoraimundia arequipensis, Borzicactus sp., Islaya mollendoensis, Trichocereus sp., Neoporteria islayensis, Tephrocactus sp., Pilocereus sp. Neoraimundia arequipensis, Borzicactus sp., Islaya mollendoensis, Tichocereus sp., Neoporteria islayensis, Tephrocactus sp. Eragrostis peruviana, Cotula australis, Tillandsia sp., Poa sp., Urocarpidium sp., Atriplex sp. Phylla nodiflora, Citharexylum flexuosum, Grindelia glutinosa, Croton ruizianus, Heliotropium lanceolatum, Vigueria weberbaueri, Lycopersicum peruvianum, Urocarpidium peruvianum, Cotula australis Caesalpinia spinosa, Duranta armata, Heliotropium arborescens Caesalpinia spinosa, Duranta armata, Heliotropium arborescens, Phylla nodiflora, Citharexylum flexuosum, Grindelia glutinosa, Croton ruizianus, Heliotropium lanceolatum, Vigueria weberbaueri, Lycopersicum peruvianum

Shrubs

620 850

20 35

Trees (cover <60%) Shrubs or trees (cover >60%)

620 690 690 980

10 50 75 100

phosphorus was extracted using an anion exchange resin (Resin-P) (Amer et al., 1955). Sodium bicarbonate 0.5 M (pH 8.5) removed labile Pi (NaHCO3-Pi) and Po (NaHCO3Po) sorbed to the soil surfaces (Bowman and Cole, 1978a,b). NaHCO3-Po is easily mineralizable and can contribute to plant available P. Sodium hydroxide 0.1 M extracted Pi (NaOH-Pi) associated with amorphous and some crystalline Al and Fe oxides (Syers et al., 1969) and Po associated with humic compounds (NaOH-Po) (Fares et al., 1974). NaOH-Pi is relatively labile Pi (Bowman and Cole, 1978a,b) while NaOH-Po is considered to be involved in long term transformation of soil under temperate climates (Tiessen et al., 1983). Resin-P, NaHCO3-Pi, NaOH-Pi, NaHCO3-Po and NaOH-Po are considered as non-occluded forms (Walker and Syers, 1976). Phosphorus extracted with 1M hydrochloric acid (HCl-P) is mainly apatitic phosphorus. It is unavailable in the short term. The residue containing the most chemically stable Po and Pi forms was digested using concentrated H2SO4 + H2O2 (Resid-P) (Thomas et al., 1967). Extracts containing organic phosphorus were digested for total P determination using a persulfate digestion method (Standard Methods, 1971). Phosphorus in the extracts or digests was determined after pH adjustment if necessary, using the ascorbic acid molybdenum blue method. A literature review of the Hedley P fractionation method was performed by Cross and Schlesinger (1995). All the chemical results were expressed on air dried basis. 2.4. Statistical methods All the statistical analyses were performed using Systat 8.0 software. Analysis of variance was used to compare P

contents between landscape types. When global ANOVA p value was < 0.05, the Bonferroni post hoc test was performed to determine which pairs of means differ significantly.

3. Results 3.1. General characteristics The mean pH values (Table 2) were not significantly different between stands. They were very acid (pH around 4.7). All the soils were very poor in clay (range 2.1 12.7%) and presented large variability concerning silt and sand (range 11.0 60.8 and 26.5 87.0%, respectively). The soils were classified as loamy sand, sandy loam or silt loam. Cac stands were very rich in sand content (87.0%) and very poor in clay (2.1%). Inversely, Tree and especially ShrTree were the richest in clay content (11.2% and 12.7%, respectively). Organic carbon contents were under 1% in Cac and CacHerb. Then they increased in Shr and Tree (1.45% and 2.40%, respectively) and above all in the ShrTree stands (6.50%). 3.2. Phosphorus contents Considering the sum of the fractions (Table 3), Cac presented the lowest value (448.1 Ag g- 1) and ShrTree the highest value (962.8 Ag g 1). The three other stands were not significantly different ( P > 0.05). Resin-P varied from 44.7 Ag g 1 in the Cac stands to 104.5 Ag g 1 in the ShrTree stands. These values are significantly different from

Table 2 General characteristics of the soil for each Lomas type (4 replicates in each Lomas type) Lomas type Cacti Cacti and herbaceous Shrubs Trees (cover <60%) Shrubs and trees (cover >60%) pH 4.9 T 0.21 4.5 T 0.19 5.0 T 0.19 4.6 T 0.15 4.7 T 0.27 % Organic C 0.34 T 0.28 0.68 T 0.16 1.45 T 0.38 2.40 T 0.47 6.50 T 0.42 % Clay 2.1 T 0.71 6.1 T 0.37 7.5 T 0.33 11.2 T 0.85 12.7 T 0.83 % Silt 11.0 T 3.42 39.0 T 3.64 59.2 T 1.07 59.3T 60.8 T 2.08 % Sand 87.0 T 4.10 54.9 T 3.97 33.4 T 1.38 29.8 T 2.49 26.5 T 2.87 Soil textural classes Loamy sand Loamy sand sandy loam Silt loam Silt loam Silt loam

A. Fabre et al. / Catena 65 (2006) 80 86 Table 3 Concentration of forms of phosphorus (Ag g 1) Lomas type Cacti Cacti and herbaceous Shrubs Trees (cover <60%) Shrubs or trees (cover >60%) Resin-P 44.7 T 4.6 (10.0) 78.1 T 6.1 (12.5) 78.9 T 6.2 (9.9) 81.6 T 6.7 (10.8) 104.5 T 2.9 (10.9) NaHCO3-Pi 23.4 T 4.3 (5.2) 33.4 T 2.7 (5.3) 122.9 T 4.3 (15.5) 109.9 T 7.9 (14.5) 92.3 T 7.1 (9.6) NaHCO3-Po 10.4 T 2.5 (2.3) 76.6 T 6.0 (12.3) 62.4 T 3.4 (7.8) 84.9 T 5.3 (11.2) 122.4 T 3.5 (12.7) NaOH-Pi 33.4 T 5.6 (7.5) 72.0 T 11.5 (11.5) 70.9 T 4.8 (8.9) 94.8 T 7.0 (12.5) 125.1 T13.2 (13.0) NaOH-Po 22.1 T 4.9 (4.9) 9.8 T 1.9 (1.6) 27.4 T 5.5 (3.4) 49.9 T 4.8 (6.6) 139.3 T 4.3 (14.5) HCl-P 267.3 T 5.1 (59.7) 297.8 T 2.7 (47.7) 386.3 T 35.2 (48.6) 256.4 T 27.8 (33.9) 136.4 T 16.3 (14.2) Residual-P 47.0 T 3.9 (10.5) 57.2 T 10.8 (9.2) 46.2 T 7.7 (5.8) 79.8 T 6.2 (10.5) 242.8 T 10.4 (25.2)

83

Sum of the fractions 448.1 T13.0 624.8 T 23.1 795.0 T 39.4 757.2 T 53.3 962.8 T 40.1

Mean values and standard error of the mean (n = 4). In brackets, percentage of the sum of the fractions.

Resin-P contents in the other stands which did not present significant differences between each other. NaHCO3-Pi contents were lowest in Cac or CacHerb (23.4 and 33.4 Ag g 1, respectively) and differed significantly with ShrTree, Tree and Shr (92.3, 109.9 and 122.9 Ag g 1, respectively), themselves being not significantly different. NaHCO3-Po varied from 10.4 (Cac) to 122.4 Ag g 1 (ShrTree). These contents were significantly different from the three other stands which were not significantly different between each other (range 62.4 84.9 Ag g 1). NaOH-Pi contents were significantly different between Cac (33.4 Ag g 1) and ShrTree (125.1 Ag g 1). The three other stands presented intermediate values (range 70.9 94.8 Ag g 1). NaOH-Po presented the lowest values in Cac, CacHerb and Shr (not significantly different; range 9.8 27.4 Ag g 1) contrasting to the content in ShrTree (139.3 Ag g 1). The content in Tree (49.9 Ag g 1) was significantly different from the other stands. HCl-P opposed a low content in ShrTree (136.4 Ag g 1) to the other stands (range 256.4 386.3 Ag g 1). ResidP markedly opposed ShrTree stand (242.8 Ag g 1) to the other stands (range 47.0 79.8 Ag g 1). 3.3. Relations between soil parameters Organic carbon was positively correlated (r = 0.90) with %cover (Table 4). Except on HCl-P and Resid-P, all the other forms of phosphorus are significantly positively correlated

with %clay or %silt or both, and negatively correlated with %sand. Likewise, except on NaHCO3-Pi or HCl-P, all the P forms were positively correlated with %cover.

4. Discussion 4.1. Carbon content The high correlation between organic carbon and vegetation cover has already been shown in several studies in arid areas (Le Houerou, 1986; Gauquelin et al., 1998). Nevertheless we can notice the high organic carbon content (around 6.50%) of the soils of the ShrTree stands, generally situated in the upper part of the Lomas, where the percentage of plant coverage is high (> 75%). 4.2. Phosphorus content In all the soils, we found high labile P contents (Resin-P, NaHCO3-Pi and Po), in comparison with data from other arid or desert soils. Nevertheless, comparisons with literature data are difficult because most of these data concern hot arid areas or deserts not periodically exposed to intense rainy periods (ENSO events). The high concentrations of the different forms of P in the Lomas can be ascribed to the combination of different and independent effects (Fig. 2).

Table 4 Correlation matrix between phosphorus forms and related parameters (in bold character: statistical significance at P < 0.05 level) Resin P Resin P NaHCO3 Pi NaHCO3 Po NaOH-Pi NaOH-Po HCl-P Residual P % Organic C Altitude % Cover % Clay % Silt % Sand 0.53 0.89 0.78 0.65 0.33 0.65 0.68 0.73 0.75 0.82 0.77 0.79 NaHCO3 Pi NaHCO3 Po NaOH Pi NaOH Po HCl P Residual P % Organic C Altitude % Cover % Clay % Silt

0.50 0.55 0.36 0.08 0.24 0.38 0.47 0.51 0.66 0.83 0.81

0.87 0.71 0.47 0.74 0.76 0.73 0.73 0.88 0.81 0.83

0.74 0.41 0.79 0.76 0.70 0.79 0.89 0.76 0.79

0.72 0.96 0.96 0.55 0.93 0.75 0.51 0.56

0.73 0.67 0.06 0.60 0.41 0.08 0.14

0.93 0.51 0.91 0.71 0.44 0.49

0.66 0.90 0.80 0.61 0.65

0.61 0.78 0.81 0.82

0.80 0.62 0.66

0.88 0.92

0.99

84

A. Fabre et al. / Catena 65 (2006) 80 86

Fig. 2. Mechanism of distribution of soil phosphorus in the Lomas de Arequipa: flowchart.

4.2.1. Land use and grazing effect Since the Spanish colonization, the Lomas has been grazed by sheep, goats and cattle. Nowadays, the Lomas are still grazed, especially during ENSO events and are used as a fuelwood source. Livestock foraging is important in pasture nutrient cycling because they convert nutrients from unavailable forms (natural fodder) to available forms (excreta) (Buschbacher, 1987). Moreover, the constant movement of the animals leads to a relatively regular distribution of faeces through the patchy landscape (Turner, 1998). 4.2.2. ENSO events During ENSO events, seeds lying within the soil, germinate and emerge into a continuous blanket. Then, this vegetation dies and decays quickly. In temperate or tropical ecosystems, many studies have shown that the different forms of P, and especially the more labile, present seasonal fluctuations. Generally, the more labile forms of P increase during winter and decrease during the growing season (Timmons et al., 1970; Saunders and Metson, 1971; Dormaar, 1972; Vaughan et al., 1986; Sarathchandra et al., 1989; Perrott et al., 1990; Magid and Nielsen, 1992). Likewise, in a mature tropical moist forest, inorganic P peaks during the dry season (Yavitt and Wright, 1996). These findings suggest that during the dormant vegetation

season (winter or dry season) there is an increase and accumulation of the more labile P forms. Considering the mechanism, the literature yields conflicting reports. Some authors consider that accumulation of labile P results on the microbial mineralization of plant debris or to the release of Pi from the organic matter (Saunders and Metson, 1971). Others attribute the labile P increase to the microbial biomass killed by air-drying (Srivastava, 1997). Using New Zealand acid soils, Haynes and Swift (1985) showed that drying soils increased phosphate extractable with EDTA, resin or NaHCO3 and considered that drying soil conducive to the release of P associated with organic matterFe and Al complexes, and possibly from killed microbial cells. Similarly, Sparling et al. (1985), studying 18 pasture soil samples from New Zealand, showed that, in most of the soils investigated, drying led to an increase of NaHCO3-Pi. Williams (1996) showed that a greater concentration of P leached by CaCl2, extracted from spruce or pines humus, coincided with drying of the soil during summer. He considered that the enhanced Pi contents in the dried soils can be mainly accounted for by the release of Pi from the killed cells or to death of fine roots and microorganisms and concluded that a rainy period following a dry period, could contribute to plant growth following rewetting. In this respect a period of soil drying could benefit overall fertility levels.

A. Fabre et al. / Catena 65 (2006) 80 86

85

Foliar or plant residue or litter leaching is generally considered as a source of labile-P (Timmons et al., 1970; Bromfield and Jones, 1972; Duffy et al., 1985; Johnson and Todd, 1987; Weiss et al., 1991; Polglase et al., 1992). In the Lomas, at the end ENSO-related rainfall, and during the beginning of the dry period, death and decay of the vegetation (especially the ephemerals), possibly causes the release and accumulation of labile P permitting the P pool to be rebuilt. During the dry period, this pool is not used by the seeds and the vegetation is dormant. 4.2.3. Particle size distribution The distribution of the vegetation from Cac to ShrTree from near 160 to 980 m of altitude can be considered as a toposequence. Generally, in a toposequence, erosional processes bring about enrichment in fine particles from the top to the bottom of the relief. In the Lomas, we found the reserve with a higher fine particles content in soil from the upper part of the landscape (Table 2). This can be ascribed to the increasing percentage plant cover from Cac (lower part) to TreeShr stands (upper part) where canopy and litter reduce erosion processes. The result is an increasing content of P labile forms from the lower to the upper part of the landscape corresponding to the general association between labile P and the finest soil particles. A positive relation between the finest soil particles and labile P was shown in cultivated and uncultivated soils (Tiessen et al., 1983) or with algal available P or P sorption in eutrophication studies (Syers et al., 1969; Dorich et al., 1984; Keulder, 1982). Nevertheless, in a toposequence from semiarid northeastern Brazil, Agbenin and Tiessen (1995) found a downslope decreasing total P concentration as in the Lomas. They concluded from the studied toposequence that in arid environments, the distribution of P results from complex interactions of lithology, weathering, colluvial actions and climatic conditions (moisture deficit followed by intense rainfalls). In the Lomas, the plant cover increases from the bottom to the top of the relief where heavy fogs (May to October) enable vegetation growth, limiting erosion processes at the top with subsequent accumulation of the different forms of phosphorus generally associated to finest soil particles.

Considering phosphorus, two periods are particularly important: a) the beginning of the drought with release of labile P (plants decaying and microbial cells killed) and its accumulation in the soil; b) rainfall with wetting of the soil permitting the growth of vegetation, especially of the ephemeral burst in a continuous blanket. In this respect the Lomas characteristics, that is the rainy period following the long dry period contribute to the overall fertility of the soil, especially as the labile forms of phosphorus are concerned the most.

Acknowledgments The authors thank M.F. Bellan and D. Lacaze for the field assistance and F. Barthelat, K. Saint-Hilaire and M. Saurat for help with many chemical analyses. The study received financial support from European Communities: Contrat U.E. n- TS3 CT 94 0324 (1995 1998): Fog as a new water resource for the sustainable development of the ecosystem of the Peruvian and Chilean coastal desert. Project Coordinator: Dr Roberto Semenzato (1995 1997) and Dr Mario Falciai (1997 1998).

References
Agbenin, J.O., Tiessen, H., 1995. Phosphorus forms in particle-size fractions of a toposequence from northeast Brazil. Soil Sci. Soc. Am. J. 59, 1687 1693. Amer, F., Bouldin, D.R., Black, C.A., Duke, F.R., 1955. Characterization of soil phosphorus by anion exchange resin adsorption and P32 equilibration. Plant Soil 6, 391 408. Bowman, R.A., Cole, C.V., 1978a. Transformations of organic phosphorus substances in soils as evaluated by NaHCO3 extraction. Soil Sci. 125, 49 54. Bowman, R.A., Cole, C.V., 1978b. An exploratory method for fractionation of organic phosphorus from grassland soils. Soil Sci. 125, 95 101. Bromfield, S.M., Jones, O.L., 1972. The initial leaching of hayed-off pasture plants in relation to the cycling of phosphorus. Aust. J. Agr. Res. 23, 811 824. Buschbacher, R.J., 1987. Cattle productivity and nutrient fluxes on an Amazon pasture. Biotropica 19, 200 207. Cross, A.F., Schlesinger, W.H., 1995. A literature review and evaluation of the Hedley fractionation: applications to the biogeochemical cycle of soil phosphorus in natural ecosystems. Geoderma 64, 197 214. Cross, A.F., Schlesinger, W.H., 2001. Biological and geochemical controls on phosphorus fractions in semiarid soils. Biogeochemistry 52, 155 172. Dorich, R.A., Nelson, D.W., Sommers, L.E., 1984. Availability of phosphorus to algae from eroded soil fractions. Agric. Ecosyst. Environ. 11, 253 264. Dormaar, J.F., 1972. Seasonal pattern of soil organic phosphorus. Can. J. Soil Sci. 52, 107 112. Duffy, P.D., Schreiber, J.D., McDowell, L.L., 1985. Leaching of nitrogen, phosphorus, and total organic carbon from loblolly pine litter by simulated rainfall. Forest Sci. 31, 750 759. Fares, F., Fardeau, J.C., Jacquin, F., 1974. Etude quantitative du phosphore organique dans differents types de sols. Phosphore Agric. 63, 25 41. Ferreyra, R., 1977. Endangered species and plant communities in Andean and coastal Peru. In: Prance, G.T., Elias, T.S. (Eds.), Extinction is Forever. Botanical Garden, Bronx, New York, pp. 150 157.

5. Conclusion The Lomas constitute an original landscape chiefly characterized by: a) the localization near the Pacific Ocean and the presence of the cold Humboldt Current; b) the topography (regular increase of the altitude from around 100 to 1000 m) which acts as a barrier to Ocean influences, causing fogs, especially between 600 to 1000 m (May to October) or receiving heavy rainfalls during ENSO events. With regard to the vegetation, the specific climatic conditions leads to a strong contrast between long periods of seed dormancy then short periods of growth, the trigger mechanism being rainfalls associated to ENSO events.

86

A. Fabre et al. / Catena 65 (2006) 80 86 Saunders, W.M.H., Metson, A.J., 1971. Seasonal variation of phosphorus in soil and pasture. N.Z. J. Agric. Res. 14, 307 328. Sparling, G.P., Whale, K.N., Ramsay, A.J., 1985. Quantifying the contribution from the soil microbial biomass to the extractable P levels of fresh and air-dried soils. Aust. J. Soil Res. 23, 613 621. Srivastava, S.C., 1997. Microbial contribution to extractable N and P after air-drying of dry tropical soils. Biol. Fert. Soils 26, 31 34. Standard Methods for the Examination of Water and Wastewater, 1971. Thirteenth ednR American Public Health Association, Washington. Syers, J.K., Shah, R., Walker, T.W., 1969. Factionation of phosphorus in two alluvial soils and particle size separate. Soil Sci. 108, 283 289. Thomas, R.L., Sheard, R.W., Moyer, J.R., 1967. Comparison of conventional and automated procedures for nitrogen, phosphorus, and potassium analysis of plant material using a single digestion. Agron. J. 59, 240 246. Tiessen, H., Stewart, J.W.B., Moir, J.O., 1983. Changes in organic and inorganic phosphorus composition of two grassland soils and their particle size fractions during 60 90 years of cultivation. J. Soil Sci. 34, 815 823. Timmons, D.R., Holt, R.F., Latterell, J.J., 1970. Leaching of crop residues as a source of nutrients in surface runoff water. Water Resour. Res. 6, 1367 1375. Turner, M.D., 1998. Long-term effects of daily grazing orbits on nutrient availability in Sahelian West Africa: I. Gradients in the chemical composition of rangeland soils and vegetation. J. Biogeogr. 25, 669 682. Vaughan, C.E., Center, D.M., Jones, M.B., 1986. Seasonal fluctuations in nutrient availability in some northern California annual range soils. Soil Sci. 141, 43 51. Walker, T.W., Syers, J.K., 1976. The fate of phosphorus during pedogenesis. Geoderma 15, 1 19. Weiss, D.M., Tremolieres, M., Carbinier, R., 1991. Biodisponibilite ` comparee du phosphore en fonction des substrats et de la frequence des inondations dans trois forets rhenanes de la plaine dAlsace. C. R. Acad. Sci., Paris 313, 245 251. Williams, B.L., 1996. Total, organic and extractable-P in humus and soil beneath Sitka spruce planted in pure stands and in mixture with Scots pine. Plant Soil 182, 177 183. Yavitt, J.B., Wright, S.J., 1996. Temporal patterns of soil nutrients in a Panamanian moist forest revealed by ion-exchange resin and experimental irrigation. Plant Soil 183, 117 129. Zavala Yupanqui, L., 1993. La sequia: causas y consecuencias en el Peru. Bol. Lima 90, 21 31.

Ferreyra, R., 1993. Registros de la vegetacion en la costa peruana en relacion con el fenomeno El Nino. Bull. Inst. Fr. Etudes Andines 22 (1), 259 266. Gauquelin, T., Jalut, G., Iglesias, M., Valle, F., 1998. Phytomass and carbon storage in the steppes of Eastern Andalusia, Spain. Ambio 27, 99 103. Haynes, R.J., Swift, R.S., 1985. Effects of air drying on the adsorption and desorption of phosphate and levels of extractable phosphate in a group of acid soils, New Zealand. Geoderma 35, 145 157. Hedley, M.J., Stewart, J.W.B., Chauhan, B.S., 1982. Changes in inorganic soil phosphorus fractions induced by cultivation practices and by laboratory incubations. Soil Sci. Soc. Am. J. 46, 970 976. Johnson, D.W., Todd, D.E., 1987. Nutrient export by leaching and wholetree harvesting in a loblolly pine and mixed oak forest. Plant Soil 102, 99 109. Keulder, P.C., 1982. Particle size distribution and chemical parameters of the sediments of a shallow turbid impoundment. Hydrobiologia 91, 341 353. Lajtha, K., 1988. The use of ion-exchange bags for measuring nutrient availability in an arid ecosystem. Plant Soil 105, 105 111. Lajtha, K., Schlesinger, W.H., 1988. The biogeochemistry of phosphorus cycling and phosphorus availability along a desert soil chronosequence. Ecology 69, 24 39. Le Houerou, H.N., 1986. The deserts and arid zones of Northern Africa. In: Evenari, H.N., et al. (Eds.), Hot Deserts and Arid Shrublands. Elsevier Science Publisher, Amsterdam, pp. 101 147. Magid, J., Nielsen, N.E., 1992. Seasonal variation in organic and inorganic phosphorus fractions of temperate-climate sandy soils. Plant Soil 144, 155 165. Osses McIntyre, P., 1996. Evaluation of the potential of the fog for water collection in the coastal area of Arequipa. Workshop, Evaluation of Fog as a Water Resource. European Commission. Programme STD3, September 30 October 4, 1996, Arequipa, Peru, pp. 1 20. Pefaur, J., 1982. Dynamics of plant communities in the Lomas of Southern Peru. Vegetatio 49, 163 171. Perrott, K.W., Sarathchandra, S.U., Waller, J.E., 1990. Seasonal storage and release of phosphorus and potassium by organic matter and the microbial biomass in a high-producing pastoral soil. Aust. J. Soil Res. 28, 593 608. Polglase, P.J., Comeford, N.B., Jokela, E.J., 1992. Leaching of inorganic phosphorus from litter of Southern pine plantations. Soil Sci. Soc. Am. J. 56, 573 577. Sarathchandra, S.U., Perrott, K.W., Littler, R.A., 1989. Soil microbial biomass: influence of simulated temperature changes on size, activity and nutrient content. Soil Biol. Biochem. 21, 987 993.